Temperature detection method, device, electronic device, and computer-readable storage medium

Abstract

The present invention discloses a temperature detection method, device, electronic device, and computer-readable storage medium. The method comprises: obtaining a feature map corresponding to an infrared thermal imaging image, wherein the infrared imaging image includes a target device; determining a local image containing a heating portion of the target device using a region proposal network based on the feature map; determining a position-sensitive score map based on the local image; and determining the local temperature of the heating portion of the target device based on the position-sensitive score map. The present invention solves the technical problem in related arts of the difficulty in accurately detecting potentially excessively high local temperatures of equipment in a computer room when detecting the temperature of the computer room.

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to a temperature detection method, device, electronic equipment, and computer-readable storage medium. Background Art

[0002] Currently, virtually every computer room has implemented environmental monitoring, equipped with numerous sensors, temperature and humidity monitors, and other tools. However, safety incidents (such as fires and power outages) still appear in local news reports. The root cause is that these temperature and humidity monitors primarily measure the average temperature of the entire room and fail to accurately detect temperature fluctuations within individual equipment components. Consequently, localized overheating can occur, often undetected by the monitors and resulting in potential safety hazards.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] Embodiments of the present invention provide a temperature detection method, device, electronic device, and computer-readable storage medium to at least solve the technical problem in related technologies that it is difficult to accurately detect local excessive temperatures of equipment in a computer room when detecting the temperature of the computer room.

[0005] According to one aspect of an embodiment of the present invention, a temperature detection method is provided, comprising: obtaining a feature map corresponding to an infrared thermal imaging image, wherein the infrared imaging image includes a target device; based on the feature map, determining a local image containing a heating part of the target device using a region proposal network; determining a position-sensitive score map based on the local image; and determining the location temperature of the heating part of the target device based on the position-sensitive score map.

[0006] Optionally, based on the feature map, a region proposal network is used to determine a local image containing the heating part of the target device, including: based on a clustering algorithm, determining a predetermined frame size for selecting the heating part of the target device; according to the predetermined frame size, selecting the feature map to obtain multiple groups of framed areas; and using a region proposal network to determine a local image containing the heating part of the target device from the multiple groups of framed areas.

[0007] Optionally, determining the position sensitive score map based on the local image includes: translating and reducing the local image to obtain a target image, wherein the proportion of the heating parts of the target device included in the target image is greater than a predetermined proportion; and determining the position sensitive score map based on the target image.

[0008] Optionally, determining the location temperature of the heating part of the target device based on the location-sensitive score map includes: determining the location temperature of the heating part of the target device as multiple confidence levels of multiple temperatures based on the location-sensitive score map, wherein the multiple temperatures correspond one-to-one to the multiple confidence levels; and determining the location temperature of the heating part of the target device based on the multiple confidence levels.

[0009] Optionally, it also includes: determining the location of the heating part of the target device; determining the location temperature of the heating part of the target device at the location at a predetermined period; and determining the location temperature curve of the heating part of the target device at the location.

[0010] Optionally, the method further includes: sending an alarm message to a predetermined terminal when the temperature of the part exceeds a predetermined temperature threshold.

[0011] According to one aspect of an embodiment of the present invention, a temperature detection method is provided, including: an acquisition module for acquiring a feature map corresponding to an infrared thermal imaging image, wherein the infrared imaging image includes a target device; a first determination module for determining, based on the feature map, a local image containing a heating part of the target device using a region proposal network; a second determination module for determining a position-sensitive score map based on the local image; and a third determination module for determining the part temperature of the heating part of the target device based on the position-sensitive score map.

[0012] According to one aspect of an embodiment of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement any of the above-mentioned temperature detection methods.

[0013] According to one aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform any of the above-mentioned temperature detection methods.

[0014] According to one aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the temperature detection method described above is implemented.

[0015] In an embodiment of the present invention, a feature map corresponding to an infrared thermal imaging image including a target device is obtained, and based on the feature map, a region proposal network is used to determine a local image including a heating part of the target device, and then the temperature of the heating part of the target device in the local image is determined. That is, a position sensitive score map can be determined based on the local image, and the local temperature of the heating part of the target device can be determined based on the obtained position sensitive score map. Because the local image of the heating part of the target device is determined, and then the local temperature of the heating part of the target device in the local image is determined based on the position sensitive score map, the technical problem in the related art that it is difficult to accurately detect the possible excessive local temperature of the equipment in the computer room when detecting the temperature of the computer room is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1 is a flow chart of a temperature detection method according to an embodiment of the present invention;

[0018] Figure 2 is a flow chart of a method for detecting temperature in a computer room provided by an optional embodiment of the present invention;

[0019] Figure 3 is a structural block diagram of a temperature detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretations:

[0023] Region Proposal Network: Region Proposal Network, referred to as RPN, takes an image (of any size) as input and outputs a set of rectangular target proposal boxes.

[0024] Residual Networks (ResNets) normalize input data and intermediate layer data. This approach allows the network to use stochastic gradient descent (SGD) during backpropagation. They are easy to optimize and can improve accuracy by increasing depth. The residual blocks within them use skip connections, allowing the network to converge, addressing the issue of exploding and vanishing gradients caused by deepening the network.

[0025] R-FCN: A network that leverages FCN to achieve greater network parameter and feature sharing and address the positional sensitivity issues of fully convolutional networks. The original image undergoes convolution to produce feature map 1. One subnetwork, similar to FastRCNN, uses the RPN to slide over feature map 1 to generate region proposals for future use. Another subnetwork continues the convolution to produce k^2 (k=3) deep feature maps 2. Based on the RoI (region proposal) generated by the RPN, these feature maps 2 are pooled, scored, and classified to produce the final detection result.

[0026] Example 1

[0027] According to an embodiment of the present invention, an embodiment of a temperature detection method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] Figure 1 FIG. 1 is a flow chart of a temperature detection method according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0029] Step S102, obtaining a feature map corresponding to an infrared thermal imaging image, wherein the infrared imaging image includes a target device;

[0030] Step S104: Based on the feature map, a region proposal network is used to determine a local image containing a heating part of the target device;

[0031] Step S106, determining a position sensitive score map based on the local image;

[0032] Step S108 : determining the location temperature of the heating location of the target device according to the location sensitivity score map.

[0033] Through the above steps, by obtaining a feature map corresponding to the infrared thermal imaging image including the target device, a local image containing the heating part of the target device is determined based on the feature map using a region proposal network, and then the temperature of the heating part of the target device in the local image is determined. That is, a position sensitive score map can be determined based on the local image, and the local temperature of the heating part of the target device can be determined based on the obtained position sensitive score map. Because the local image of the heating part of the target device is determined, and then the local temperature of the heating part of the target device in the local image is determined based on the position sensitive score map, the technical problem in the related technology that it is difficult to accurately detect the local excessive temperature of the equipment in the computer room when detecting the temperature of the computer room is solved.

[0034] As an optional embodiment, an infrared thermal imaging image including a target device is obtained, wherein the infrared thermal imaging image may be an image obtained by a thermal imaging system. Then, based on the infrared thermal imaging image, a feature map corresponding to the infrared thermal imaging image is obtained. Before obtaining the feature map corresponding to the infrared thermal imaging image based on the infrared thermal imaging image, the infrared thermal imaging image may be preprocessed. Because the infrared thermal imaging image converts the infrared band signal of the thermal radiation of the object into an image and graphics that can be visually distinguished, the image is easily affected by the ambient temperature, light, etc. Therefore, the brightness of the three components of the image can be used as the grayscale value of the pixel to perform a grayscale processing operation on the image. Secondly, the image can also be sharpened to reduce the blur in the image and enhance the image edges, making the image easier to process and the obtained feature map more accurate.

[0035] As an optional embodiment, a feature map corresponding to an infrared thermal imaging image is obtained. There are many ways to obtain a feature map corresponding to an infrared thermal imaging image, such as residual networks and scale-invariant feature transformation. Preferably, a ResNet residual network can be used to extract features from the infrared thermal imaging image to obtain a feature map. Using a residual network to obtain a feature map from an infrared thermal imaging image can shorten the time required to obtain the feature map and improve the quality of the obtained feature map. The residual network simplifies the optimization process by adding an identity mapping, decomposing a problem into residual problems at multiple scales, thereby optimizing training. Furthermore, a shortcut connection is added to the residual network. This shortcut connection allows data to be processed across layers for nonlinear superposition without adding additional parameters or computational overhead to the network, thereby improving network training speed and effectiveness. This simple structure effectively addresses degradation issues as the model's layers increase. Specifically, by using skip connections, the network converges, addressing the issues of exploding and vanishing gradients that can occur with a deeper network.

[0036] As an optional embodiment, based on the feature map, a region proposal network is used to determine a local image containing the heating part of the target device. Through the region proposal network, the heating part of the target device in the feature map can be accurately located, and the target can be located quickly and accurately. It should be noted that there can be many heating parts of the target device, for example, multiple heating parts such as the CPU, hard disk, etc., and the heating parts should not be the same for different target devices. Therefore, it is necessary to customize the heating parts of the target device according to the actual application and scenario. The local image containing the heating part of the target device can be determined based on the color in the infrared thermal imaging image and the feature map. The ability to identify the local part of the equipment in the computer room solves the technical problem in the related art that it is difficult to detect the temperature of the local part of the equipment in the computer room when detecting the temperature of the computer room.

[0037] As an optional embodiment, when a region proposal network is used to determine a local image containing the heating part of a target device based on a feature map, a predetermined frame size for selecting the heating part of the target device can be first determined based on a clustering algorithm. The predetermined frame size can be obtained based on the target device, that is, a K-means clustering algorithm based on intersection-over-union can be used to learn the prior knowledge of the frame size of the heating part of the target device from a large amount of training data to obtain a predetermined frame size suitable for this application. This allows the feature map to be framed based on the predetermined frame size to obtain multiple groups of framed regions, and then the region proposal network can be used to determine the local image containing the heating part of the target device from the multiple groups of framed regions, thereby improving efficiency.

[0038] As an optional embodiment, a position sensitive score map is determined based on a local image. By means of the position sensitive score map, the temperature of the heating part of the target device in the local image can be accurately determined. Optionally, in the process of determining the position sensitive score map based on the local image, because the size of the heating part of the target device in the local image is different, it is impossible to fully adapt to the size of the predetermined frame selection size. Therefore, when determining the position sensitive score map based on the local image, the local image can also be adjusted to obtain the position sensitive score map. For example, the local image can be translated and reduced to obtain a target image in which the proportion of the heating part of the target device is greater than the predetermined proportion, and then the position sensitive score map is determined based on the target image. Through the above operation, the target image can include a larger proportion of the heating part of the target device than the local image, reducing the influence of the background area (non-target device heating part). The score obtained by the position sensitive score map can be more accurate, thereby more accurately determining the temperature of the heating part of the target device.

[0039] It should be noted that the step of determining the position-sensitive score map can be processed by a convolutional layer, which can be placed at the end of the region proposal network. That is, based on the feature map, the region proposal network is used to determine the local image containing the heating part of the target device, and based on the local image, the position-sensitive score map is determined.

[0040] As an optional embodiment, when determining the location temperature of the heating part of the target device based on the location sensitivity score map, the location temperature of the heating part of the target device can be determined as multiple confidence levels of multiple temperatures based on the location sensitivity score map, wherein the multiple temperatures correspond to the multiple confidence levels one-to-one; and the location temperature of the heating part of the target device is determined based on the multiple confidence levels. Optionally, the temperature with the highest confidence level can be directly selected as the location temperature of the heating part of the target device, or a judgment condition can be added, that is, based on the highest confidence level, the highest confidence level must be greater than a predetermined confidence level before the temperature corresponding to the confidence level is determined to be the location temperature of the heating part of the target device. This allows for more accurate determination of the location temperature of the heating part of the target device.

[0041] As an optional embodiment, the method further includes determining a curve of a heating portion of the target device, which can be determined by the following steps: determining the location of the heating portion of the target device, determining the temperature of the heating portion of the target device at the location at a predetermined period, and determining a temperature curve of the heating portion of the target device at the location. Based on the curve, characteristics related to location, time, and temperature can be learned, thereby enabling better safety maintenance of the computer room.

[0042] As an optional embodiment, if the temperature of a part exceeds a predetermined threshold, an alarm message is sent to a predetermined terminal. Specifically, when the temperature exceeds the predetermined threshold, an external sound and light alarm sounds an alarm, sending an alarm message to the predetermined terminal. A real-time high-definition camera manually verifies the target information and determines whether to activate firefighting equipment and shut down the power switch in real time. The current temperature safety is determined based on the safety temperature standards set by various devices, allowing for timely resolution of potential safety hazards.

[0043] Based on the above embodiment and optional embodiment, an optional implementation manner is provided, which is described in detail below.

[0044] An optional embodiment of the present invention provides a method for detecting computer room temperature based on infrared imaging and a target detection algorithm. This method, combined with infrared thermal imaging images and based on the R-FCN network model used in the target detection algorithm, improves the network structure, candidate box generation method, and target feature map pooling method to accurately detect abnormally high temperatures in the computer room caused by long-term operation of equipment. Figure 2This is a rough flow chart of a method for detecting temperature in a computer room provided by an optional embodiment of the present invention. Figure 2 As shown, the method provided by the optional embodiment of the present invention is described in detail below:

[0045] S1, obtaining an infrared thermal imaging image of the computer room obtained by a thermal imaging system;

[0046] It's important to note that infrared thermal imaging images can be preprocessed before input. Because infrared thermal imaging converts infrared signals from an object's thermal radiation into visually discernible images and graphics, they are susceptible to environmental temperature, lighting, and other factors. Therefore, grayscaling can be performed using the brightness of each of the three image components as the grayscale value of each pixel. Furthermore, sharpening can be performed on the image to reduce blur and enhance edges, making it easier to process before inputting it into the network.

[0047] S2, combined with the R-FCN algorithm, realizes temperature detection of each heating area of ​​each device in the computer room;

[0048] S2.1, use ResNet residual network to extract features from infrared thermal imaging images and obtain feature maps;

[0049] Using a residual network to implement the process of acquiring feature maps from infrared thermal imaging images can shorten the time it takes to obtain feature maps and improve their quality. The residual network adds an identity mapping to simplify the optimization process, decomposing a problem into residual problems at multiple scales, thereby optimizing training. Furthermore, a shortcut connection is added to the residual network. This shortcut connection allows data to flow across layers for nonlinear superposition operations without adding additional parameters or computational effort to the network, thereby improving the speed and effectiveness of network training. This simple structure effectively addresses degradation issues as the model's layers deepen. Specifically, by using skip connections, the network converges, resolving the issues of exploding and vanishing gradients caused by deepening the network.

[0050] S2.2, use the region proposal network (RPN) to generate a series of candidate regions (ROI);

[0051] The region proposal network is a network structure used to generate candidate regions (the same as the local image mentioned above) on the feature map. When determining the candidate regions, multiple groups of frame selection regions must be determined first. When obtaining multiple groups of frame selection regions, the frame selection size used for frame selection can be determined first. Due to the differences in the size and aspect ratio of the target to be measured, it is often necessary to set multiple frame selection scales. In the original RPN network, the candidate frame size is composed of three default sizes (128x128, 256x256, 512x512) and three aspect ratios (1:1, 1:2, 2:1), a total of 9 types, which are used to detect targets of different sizes. In an optional embodiment of the present invention, the recognition target is set to the heating part of the device. The overall difference in target size is not large. The candidate frame of the original size is not suitable for the application of the optional embodiment of the present invention. Therefore, the optional embodiment of the present invention selects K=6 to avoid generating candidate frames of redundant and useless sizes, and at the same time improves the method of generating candidate frames in the RPN network. Using a K-means clustering algorithm based on intersection-over-union (IoU) to learn prior knowledge about the geometric size of target bounding boxes from a large amount of training data, we generated six candidate bounding box sizes, consisting of two sizes and three aspect ratios suitable for the optional implementation of this invention. Using these candidate bounding box sizes as a reference can reduce the initial error of RPN network training, improving network training speed and target positioning accuracy.

[0052] By generating candidate frames of six sizes at each feature point, the candidate area including the heat-generating part of the device can be selected.

[0053] S2.3, construct a special convolutional layer to construct a set of position-sensitive score maps for each ROI;

[0054] Taking a candidate region as an example, assume that there are 9 sub-parts in a candidate region, and set a score for each part to represent the sensitivity score of a single type of object. For example, determine the sensitivity score of each part with a temperature of 70°. The sensitivity score can represent the proportion of the sub-part with a temperature of 70°. The sub-part is generally the (upper-left, upper-middle, upper-right, middle-left, ..., lower-left) area of ​​a single type of object.

[0055] S2.4, perform mean pooling on the position-sensitive score map and classify it;

[0056] There are C categories, which can be represented as C temperatures. Average pooling outputs the mean, and finally softmax regression is used to determine the confidence level that the region belongs to category C. This gives the confidence level that the device temperature belongs to each temperature category. The temperature with the highest confidence level is selected as the device temperature.

[0057] S3 sets a temperature threshold. When the temperature exceeds the threshold, an external sound and light alarm sounds an alarm. A real-time high-definition camera manually confirms the target information and determines whether to activate firefighting equipment and shut down the power switch in real time. The safety of the current temperature is determined based on the safety temperature standards set for various devices, and any potential safety hazards are promptly addressed.

[0058] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited by the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0059] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0060] Example 2

[0061] According to an embodiment of the present invention, a device for implementing the above temperature detection method is also provided. Figure 3 : is a structural block diagram of a temperature detection device according to an embodiment of the present invention. Figure 3 As shown, the device includes: an acquisition module 302, a first determination module 304, a second determination module 306 and a third determination module 308. The device will be described in detail below.

[0062] An acquisition module 302 is configured to acquire a feature map corresponding to an infrared thermal imaging image, wherein the infrared imaging image includes a target device; a first determination module 304 is connected to the acquisition module 302 and configured to determine, based on the feature map, a local image containing a heating portion of the target device using a region proposal network; a second determination module 306 is connected to the first determination module 304 and configured to determine a position-sensitive score map based on the local image; and a third determination module 308 is connected to the second determination module 306 and configured to determine the local temperature of the heating portion of the target device based on the position-sensitive score map.

[0063] It should be noted here that the above-mentioned acquisition module 302, first determination module 304, second determination module 306 and third determination module 308 correspond to steps S102 to S108 in implementing the temperature detection method, and the instances and application scenarios implemented by the multiple modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.

[0064] Example 3

[0065] According to another aspect of an embodiment of the present invention, an electronic device is provided, including: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute the instructions to implement any of the above-mentioned temperature detection methods.

[0066] Example 4

[0067] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above-mentioned temperature detection methods.

[0068] Example 5

[0069] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the temperature detection method described above is implemented.

[0070] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0071] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0072] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0073] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0074] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0075] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0076] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A temperature detection method, characterized in that: include: Acquire a feature map corresponding to an infrared thermal imaging image, wherein the infrared thermal imaging image includes a target device; Based on the feature map, a region proposal network is used to determine a local image containing a heating portion of the target device, wherein when determining the local image, multiple groups of frame selection regions are first determined, and when the multiple groups of frame selection regions are obtained, multiple frame selection scales are set; Determining a position-sensitive score map based on the local image; The location temperature of the heating location of the target device is determined based on the location sensitivity score map.

2. The method according to claim 1, characterized in that Determining a local image containing a heating portion of the target device using a region proposal network based on the feature map includes: Determining, based on a clustering algorithm, a predetermined frame size for selecting a heating portion of the target device; Frame the feature map according to the predetermined frame size to obtain multiple groups of framed areas; A region proposal network is used to determine a local image containing the heating part of the target device from the multiple groups of selected regions.

3. The method according to claim 1, characterized in that Determining a position-sensitive score map based on the local image includes: translating and reducing the partial image to obtain a target image, wherein a proportion of the heating portion of the target device included in the target image is greater than a predetermined proportion; The position-sensitive score map is determined based on the target image.

4. The method according to claim 1, wherein Determining the temperature of the heating part of the target device according to the position-sensitive score map includes: Determining, based on the position-sensitive score map, a plurality of confidence levels of a plurality of temperatures for a location temperature of a heating location of the target device, wherein the plurality of temperatures correspond to the plurality of confidence levels in a one-to-one manner; The temperature of the heating part of the target device is determined based on the multiple confidence levels.

5. The method according to claim 1, characterized in that Also includes: Determining the location of a heating part of the target device; determining the temperature of the heat-generating portion of the target device at the location at a predetermined period; Determine a site temperature curve of a heating site of the target device at the site location.

6. The method according to any one of claims 1 to 5, characterized in that Also includes: When the temperature of the part exceeds a predetermined temperature threshold, an alarm message is sent to a predetermined terminal.

7. A temperature detection method, characterized in that: include: an acquisition module, configured to acquire a feature map corresponding to an infrared thermal imaging image, wherein the infrared thermal imaging image includes a target device; a first determination module, configured to determine, based on the feature map, a local image containing a heating portion of the target device using a region proposal network, wherein when determining the local image, a plurality of groups of frame selection regions are first determined, and when the plurality of groups of frame selection regions are obtained, a plurality of frame selection scales are set; A second determination module is configured to determine a position-sensitive score map based on the local image; The third determination module is configured to determine the location temperature of the heating location of the target device according to the location-sensitive score map.

8. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the temperature detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the temperature detection method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the temperature detection method according to any one of claims 1 to 6 is implemented.

Images